unsloth/cogito-v2-preview-deepseek-671B-MoE-GGUF · Scored 77.14 (non-reasoning) 2bit IQ2

logs
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| Model                                                  | Dataset   | Metric          | Subset           |   Num |   Score | Cat.0   |
+========================================================+===========+=================+==================+=======+=========+=========+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | computer science |    10 |  0.8    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | math             |    10 |  0.8    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | chemistry        |    10 |  0.9    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | engineering      |    10 |  0.6    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | law              |    10 |  0.5    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | biology          |    10 |  0.9    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | health           |    10 |  0.7    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | physics          |    10 |  0.8    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | business         |    10 |  0.9    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | philosophy       |    10 |  0.8    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | economics        |    10 |  1      | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | other            |    10 |  0.8    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | psychology       |    10 |  0.8    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | history          |    10 |  0.5    | default |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-deepseek-671B-MoE-GGUF_UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | OVERALL          |   140 |  0.7714 | -       |
+--------------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
unsloth
/

cogito-v2-preview-deepseek-671B-MoE-GGUF

Scored 77.14 (non-reasoning) 2bit IQ2_M on MMLU Pro single shot
